robert h. whittaker (1920–1980): the man and his...

Robert H. Whittaker (1920-1980): The man and his work

W. E. Westman I & R. K, Peet 2 i Department of Geography, University of California, Los Angeles, CA 90024, U.S.A. 2 Department of Botany, University of North Carolina, Chapel Hill, NC27514, U.S.A.

Keywords: Communi ty ecology, Gradient analysis, History of ecology, Ordination, Species diversity, Vegetation, Whittaker

Vegetatio 48, 97-122 (1982). 0042-3106/82/0482 0097/$5.20. © Dr W. Junk Publishers, The Hague. Printed in The Netherlands

98

Abstract

R. H. Whittaker enlivened many fields within ecology, systematics and evolution with his insights. Perhaps his most significant contributions to ecology lie in the development of the theories and methods of gradient analysis. Through the verification of the individualistic hypothesis with field data from many regions, and the subsequent development and dissemination of methods for studying species distributions along continua, he helped replace the Clementsian paradigm with a Gleasonian one. His extensive field data on primary production, nutrient cycling patterns and species diversity established new standards for documentation in synecology and helped clarify the basis for site-to-site variation in these variables. Through his broad command of the ecological literature, his writings and his contact with ecologists throughout the world he fostered international understanding of the diversity of approaches to vegetation study.

Introduction

At the time of his death on October 20, 1980, Robert H. Whittaker was recognized as one of the world's leading authorities on plant synecology. During his lifetime Whittaker was a major innovator of methodologies for community analysis and a leader in marshalling field data to document patterns in the composition, productivity and diversity of land plant communities. Through literature review and synthesis, he brought clarity to such disparate fields as classification and ordination of plant communities, plant succession and climax, allelochemistry, evolution and measurement of species diversity, niche theory, and the systematics of kingdoms of organisms. He produced monographs on the vegetation patterns of several montane regions of the United States, and in the last six years of his life extended his work to Mediterranean- climate and arid regions not only of the United States, but of Israel, Australia and South Africa. Whittaker's most often cited work was his undergraduate textbook, Communities and Ecos),stems (1970c, 1975c), which not only provided an introduction to synecology for thousands of students throughout the world, but also succinctly sum- marized highly diverse literature in a way that provided new insights for many professional ecologists.

Whittaker's ideas were sometimes the subject of intense controversy during his lifetime (continuum theory, climax pattern concepts, the proposal of five kingdoms of organisms). Critical evaluation of Whittaker 's contributions is best left to specialists in his various fields of endeavor, and the ultimate significance of his ideas will have to be evaluated by future historians of ecology. As former students

and then colleagues of Whittaker, we lay no special claim to such tasks. Rather, it is our aim in this article to summarize Whittaker's major ideas and contributions, and to describe their development within the perspective of the major events in his training and career. In researching his career, we solicited insights from many of his former as- sociates including doctoral students, research col- laborators, and close professional colleagues. The letters we received have been placed in the History of Ecology archives at the University of Georgia.

A biographical sketch

Ear(v hfe

Robert Harding Whittaker was born the young- est of three children to Clive Charles and Adeline Harding Whittaker on December 27, 1920 in Wi- chita, Kansas. The family home was in Eureka, Kansas, a town of fewer than 4 000 people, located 100 km east o f Wichita in a wheat growing, livestock grazing and oil producing area of the state. Clive Whittaker taught zoology at Fairmount College (now University of Wichita), and Adeline taught English at the same institution, z Robert 's interest and skill in languages was stimulated by his mother and his interest in natural history by his father. Clive Whittaker left his teaching post near the time of Robert 's birth and engaged in speculative oil-well drilling in an effort to raise family income to the upper middle-class standard of living to which he and his wife were accustomed as children. 2

Eureka is in the midst of the gently undulating

See notes at the end of the paper.

topography of the Great Plains. Extensive wheat growing in prairie regions of Kansas with margin- ally adequate precipitation led to massive topsoil erosion during the dry period of the 1930's when Rober t was a teenager. These were also the years of the Great Depression, when concern with economic survival pruned frivolities to their sparest.

The combinat ion of influences f rom his family and the larger mid-Western society of his day served to embue Whittaker with a belief in the work ethic which he held unswervingly throughout his career. Other values nurtured by that place and era included a stoic acceptance of adversity, a sense of identification with and loyalty to the local com~ munity, and a respect for authority, parental and social. In the late 1960's when a number of these values were being questioned, Whittaker felt particularly challenged, and spoke out clearly of his values in final paragraphs of his paper to the Brookhaven Symposium on Diversity and Stability in Ecological Systems (1969c, p. 192):

As ideals comfort, pleasure, passive and commercialentertainment, insulation from adult problems, and relative freedom from real effort and discipline may have singularly unideal consequences. It seems they may impoverish the bases of growth of inner strength, sense of self, respect for work, and sense of community with others and the society and its purposes and cultural heritage. They can produce some youth who have never learned successful management of the angers endemic to the human condition, in whom this anger must direct itself toward destruction and self-destruction. The life of passive pleasure may deny the discipline of growth by which rationality and respect for others are superimposed on the easier ways of irrationality and antagonism.' This life leads in an extreme few toward an ideological and emotional solips- ism, combined with systematic hatred of restraint and authority, hatred of others' freedom, and antagonism toward understanding of complex- ity and toward the life of the mind. I describe the extreme, but do so because I believe the United States is, by self-augmenting processes now involving commerce and the communications industry, producing increasing numbers of youth with characteristics toward this extreme.

99

These views, though rarely expressed in print, were characteristic of Whittaker 's value system. The consequences of his own self-discipline, drive for hard work, and respect for 'complexi ty and . . . the life of the mind,' are seen both in the ecological methods he developed and in the kind of research he produced.

Early in his childhood, Whittaker developed a hobby of collecting butterflies. In teenage years, he would hike through the grazing lands near Eureka, observing butterflies and other remnants of nature. In summers when his family vacationed in Colorado, he would often climb to alpine meadows at dawn, to spend the day alone in the mountains, z While Whittaker rationalized his work on mountain vegetation as providing useful vegetation gradients, he also clearly enjoyed montane areas and eagerly sought excuses to study them. 2

Formal education

In 1938 Whittaker entered Washburn Municipal University in Topeka, Kansas. He received a Bachelor of Arts degree in biology there in 1942, with principal emphasis in zoology and entomo- logy. With the Second World War in full swing, he enlisted in the Army Air Force, and was stationed in England as a weather observer and forecaster until 1946. It was during his spare moments in this period that Whittaker acquainted himself with the classical works ofTansley, Clements, Warming and others. He decided to pursue ecological work upon his return to civilian life, and entered graduate school at the University of Illinois in 1946. He completed his Ph.D. 21/2 years later.

Arthur G. Vestal was the botanist and teacher who most influenced Whittaker during his graduate career at Illinois. 3,4 Perhaps Vestal's work on the vegetation of the Colorado Rockies (1914, 1917) where Whittaker spent some of his most pleasant summers, as well as Vestal's scholarly reputation, attracted Whittaker. Although Whittaker applied for graduate standing in the Department of Botany, his application was denied, apparently because of insufficient course background in botany.3, 5 Whit- taker was subsequently admitted to the Zoology Depar tment and awarded a fellowship. In March 1946 he initiated graduate studies under the direc- tion of Victor Shelford, who retired from active teaching that summer. Charles Kendeigh replaced

100

Shelford as Whittaker's adviser in September. 5 Although Whittaker worked with Kendeigh and acknowledged his debt to him, he still called Vestal his 'second adviser.'4 Whit taker was much taken by Vestal's classroom lectures, which questioned the Clementsian notions of the plant association, and discussed Henry Gleason's alternative views of individualistic species distribution. 4 It is apparent f rom later conversations that Whittaker was deeply aware of the influence Vestal had in shaping his theoretical attitudes and analytical approach. 3 In his later years at Cornell, Whittaker was obviously pleased to play a similar role for the graduate students of others.

The academic environment Whittaker entered at Illinois has been described in a short autobio- graphical sketch (1972e). That environment was one of a crossfire of views on the nature of communities. Shelford was a strong Clementsian and Kendeigh's ideas had developed from that tradition. Vestal's sympathies lay with the less popular individualistic views of Gleason. By his own account Whittaker brought with him yet a different viewpoint, the lifezone approach of Charles H. Merriam which fit so well the Colorado mountains that he knew. Recognizing the importance of this unresolved conflict, Whittaker formulated a hypothesis. 'Because of their many interactions, species should tend to evolve toward co-adapted groupings, each representing a favorable, balanced pattern of species interactions adapted to some range of environments. Between these favorable combi- nations should lie transitions of 'community-level hybridization' with less balanced and more change- able mixtures of species populations' (1972e, p. 690).

Kendeigh had been interested in possible studies in the Great Smoky Mountains and when Whit- taker indicated he would like to undertake a study in a mountain area the project was fixed. 4,5 Originally Whittaker set out to test his hypothesis by sampling foliage insects, both in many different kinds of forests, and along an elevational transect f rom deciduous to spruce-fir forest. It soon became apparent that the test would fail because of the irregularity of insect population levels and because of lack of knowledge about how the underlying plant communities related to each other and the environment. For these reasons Whittaker then undertook a study of the vegetation of the Great

Smoky Mountains. He also attempted to measure microclimatic variation, but abandoned the effort after his weather recording instruments were de- stroyed by animals 5, and substituted topographic position and aspect as indirect indicators. The field work for both the foliage insect monograph and the Smoky Mountain vegetation monograph was completed in one four month field season.

After returning to Urbana, Whittaker spent the entire academic year analyzing his data on plant population densities and preparing his dissertation on the plant ecology of the Great Smoky Moun- tains. Having acquired an academic post for the following year, he did not have time to complete analysis of the insect populations before leaving Illinois. After promising to prepare a manuscript on the insects the following year, he successfully defended a dissertation in zoology which contained no reference to animals.

What emerged from Whittaker's work was vali- dation of Gleason's hypothesis and rejection of his own; species were independently distributed along environmental gradients and the hypothesized groups of co-adapted species with parallel distributions could not be found. The significance of the result was obvious to Whittaker and to others like W. H. Camp who wrote Whittaker that his manuscript was 'probably the most important ecological paper of the present century, ' and that the method would revolutionize the fieldP However, because Whittaker's thesis was 'long, w o r d y . . , speculative and highly theoretical '5 and in part because it was 'overly aggressive in tearing down opposing theories to make way for his own '5, publication proved difficult_ Whittaker once wrote to Kendeigh in response to a suggestion that the manuscript be shortened, 'I should be appreciative if you would avoid regarding the work as another student's thesis to be edited down to article length for publication. I need not remind you that it is a pioneering work which will mark a turning point in the study of vegetation. '5 In the intervening years Whittaker's unconventional ideas and approaches were rein- forced by the simultaneous, independent work of J_ T. Curtis and his collegues and students in Wis- consin_ After much negotiating and rewriting H. J. Oosting agreed to publish Whittaker's manuscript in Ecological Monographs where it appeared in 1956a, eight years after the dissertation was completed.

Subtler political and cultural values may also have drawn Whittaker's attention to opponents of the Clementsian view. During the 1930's, the community concept became linked with political causes (see Tobey, 1981), as well as being increasingly promoted as an article of faith_ Clements and Chaney (1936, pp. 51, 52) drew a parallel between the notions of the plant communi ty as an inter- dependent superorganism, and the desirability of developing interdependence ( 'cooperation') in human societies to deal with the effects of the Dust Bowl in the Great Plains. This use of holism as a defense of federalism was foreshadowed by the writings of the South African diplomat, Jan Smuts (1926), who used examples of holism in nature to support a variety of political causes. These applica- tions of the community concept were both defended (Phillips, 1935) and attacked (Tansley, 1935, p. 299). In the midst of this imbroglio, Henry Glea- son's individualistic notions could be seen as a challenge to those of Clements on both conceptual and political fronts. Significantly, Phillips (1935, p. 227) termed Gleason an 'iconoclast' , and further noted (1935, p. 226) that many younger investiga- tors in ecology in the early 1930's favored the Gleasonian viewpoint_ Given the political climate of this ecological debate, with its undertones of inter-generational conflict, it is plausible that Whit- taker's strong sense of individualism, both political and personal, also led him to sympathize with Vestal's attack on Clements' notions, and to favor the Gleasonian hypothesis.

Whittaker 's penchant for immersing himself in new subdisciplines and synthesizing theoretical concepts for testing from the existing literature was a pattern he was to follow throughout his career. While the study of vegetational continua is one such example which he 'pursued with field data, his first paper (1951) also criticized Clements' regional climatic climax concept. Whittaker proposed that stable vegetation itself responds to site-to-site variations in environment producing gradients of climax vegetation, and that successional patterns leading to this multiplicity of endpoints must therefore also be different. These criticisms, while tied to his Great Smoky Mountains data, were also theoretical criticisms drawn from his analysis of the existing literature. Whittaker subsequently used the literature review and synthesis approach to com- ment on a variety of topics outside the realm of his

101

own field research, notably the phylogeny of the great groups of organisms (1957b, 1959, 1969b, 1977f, 1978e), and allelochemistry (1970b, f, 1971a, f). Similarly, his major literature reviews on classification of natural communities (1962a) and evolution and measurement of diversity (1972d, 1977g) became definitive works in these fields.

Career and personal development

In 1948 Whittaker was appointed instructor in the Depar tment of Zoology at Washington State College (now University) in Pullman, Washington. While at Washington State, he conducted a comparative study of vegetation on serpentine vs. quartz diorite soils in southern Oregon (1954b, c), and began his field work on the Klamath region (1961 c) and the Siskiyou Mountains of Oregon and California in particular (1960b). He also conducted a study of copepod communities of small ponds in the Columbia Basin of Southeastern Washington (1958a). While he rose to the position of Assistant Professor at Washington State by 1950, he was let go in 1951. The reasons for this decision are not known with certainty but the catalyst was a period of fiscal retrenchment. Also, the aggressiveness with which Whittaker attacked the ideas of the established professors and defended his own5 was not apparently endearing to the faculty. 2 He later admitted that he was at the time 'something of a young turk. '2

Having lost his first academic job, Whittaker turned to industry for employment. He was hired in 1951 as a Senior Scientist in the Aquatic Biology Unit of the Department of Radiological Sciences at the Hanford Laboratories of the General Electric Company in Richland, Washington, and remained for three years. With access to laboratory facilities for radio-active tracer studies, Whittaker undertook a detailed microcosm study of the movement of phosphorus in aquaria (1961b). He recognized even at this early date that movement of radio- nuclides in the environment was an important topic, and that studies of movement of nutrients would be important for understanding ecosystems.

While at the Hanford Laboratories he struck up a romance with Clara Caroline Buehl and the two were married on New Year's day, 1953. Although Clara had an M.A. in biology, her role in the marriage soon became that of wife and mother,

102

rather than of scientific collaborator. In the years of their marriage, they raised three sons: John Charles, Paul Louis, and Carl Robert.

In 1954 after sending some 150 letters of inquiry, 6 Whittaker obtained an academic job as Instructor in the Department of Biology, Brooklyn College, City University of New York. Brooklyn College was primarily an undergraduate college, with some Masters degree students. It did not have the publish-or-perishatmosphere of a major university. During his summers, Whittaker escaped the urban life of New York by returning to the Great Smoky Mountains where he initiated a several-year effort to obtain measurements of the biomass and productivity of the forest communities along an elevational gradient. Since Whittaker's interest was in the entire aboveground primary productivity, he began to develop methods for measuring production of shrubs and herbs (1962b, 1963c), and tree components in addition to trunks (1965b). The main approach he selected was the use of volumetric measurement based on growth rings. He succeeded, through a laborious set of calculations, to obtain productivity estimates for the major plant communities in the mountain range (1966a). This effort laid the ground work for the subsequent development of the dimension analysis methodology. The patience and meticulousness required to carry through such a project virtually single- handedly were enormous. While other ecologists sought alternative routes to production measurement through dynamic techniques like gas ex- change, Whittaker 's personality was such that the enormous effort and attention to detail required to complete such calculat ionsby the growth-ring and clip-and-weigh methods perhaps even appealed to him. It could be argued that he did not have the funding or training to pursue a more physiological approach. Nevertheless, that he engaged in this type of research at all is, perhaps, a clue to the nature of problems and approaches that he was drawn to.

During this period, Whittaker began collabora- tive projects related to his quest to characterize productivity gradients. With Neil Cohen and Jerry Olson at Oak Ridge National Laboratories, he applied dimension analysis to some eastern Ten- nessee trees (1963d). With William Niering of Connecticut College he began a gradient analysis and productiwty study of the vegetation of the Santa Catalina Mountains of Arizona resulting in a

series of papers, one of which (1965d) won them the 1966 Mercer Award of the Ecological Society of America for the outstanding ecological paper published in the previous two years.

By 1964, Whittaker was an Associate Professor at Brooklyn College with a substantial reputation primarily as a field-based ecologist producing detailed, data-rich studies, and as a maverick on questions of ecological theory. Among ecologists the latter reputation stemmed primarily from his challenge to three major Clementsian paradigms: the regional climatic climax theory, the notion of unique seres leading to the climatic climax, and the supraroganismal theory of plant association. Among systematists, it derived from his proposed five kingdom system of classification. Although none of these challenges arose entirely without precedent in the literature, the strength of these challenges was due to Whittaker 's skill in marshalling exhaustively both the literature and field evidence for his viewpoints, and in writing persuasively_

In 1964 George M. Woodwell persuaded Whit- taker to take a year's leave to work with him at Brookhaven National Laboratory. The partnership was electric, in part because the two men shared a commitment to high standards and hard work, and perhaps also because Woodwell's humorful and more expansive style well complemented Whit- taker's more intense one. The two developed a profound respect and fondness for each other, and collaborated successfully on eight papers on the Brookhaven oak-pine forest and surrounding vegetation f rom the point of view of surface area (1967c), biomass and production (1967d, 1968e, g, 1969e), nutrient flow (1967d), and effects of gamma irradiation on structure (1968f, 1973n) and diversity (1969e, 1971b). Their earliest paper (1966b), with W. M. Malcolm, expressed their mutual concern with the hazards posed by technology to ecosystems, and was Whittaker's first published venture into the arena of environmental concerns. During this period, Whittaker and Woodwell for- malized and computerized the dimensional analysis system (1969e, 1971c). Whittaker stayed on a second year at Brookhaven, and decided not to return to Brooklyn College. He still thought fondly of the meadows and mountains of his youth and wanted to go West, away from the press of Eastern urban life. He accepted an offer as Professor in the

Department of Populat ion and Environmental Biology at the new University of California campus at Irvine.

Whittaker stayed at Irvine two years, long enough to initiate a study of California vegetation in the San Jacinto Mountains (1976b, 1977b, c) and to begin supervision of a doctoral student, Lawrence McHargue, who studied the vegetation of the Coachella Valley. He also became acquainted with the pygmy forest region of Mendocino County, California through Hans Jenny at Berkeley. De- spite the excellent faculty assembled at Irvine, Whittaker was disenchanted at his new location by the dizzying pace of urbanization he observed in southern California at that time. 7 The wildlands he had hoped to find were disappearing from view both by smog and bulldozer.

Disillusioned with California, Whittaker accepted an invitation from Lamont Cole to move to Cornell University in rural Ithaca in September 1968 as Professor of Biology in the Section of Ecology and Systematics. With the recognition that came with his appointment to Cornell, Whittaker began to expand the scope of his career objectives, to consolidate his theoretical positions in the literature, and to become adviser and helper to graduate students and colleagues.

Whittaker 's first graduate student at Cornell, Walter Westman, undertook a study of production, nutrient circulation and ordination of the vegetation gradient from pygmy forest through Bishop pine forest to redwoods in the Mendocino County area that Whittaker had earlier secured funding to study. Whittaker spent the summer of 1969 in Mendocino with his family partly working with Westman and his field assistants, but mostly preparing the first edition of his undergraduate text, Communities and Ecosystems (1970c). Upon return to Cornell, Hugh G. Gauch, Jr., a Cornell graduate with an M_A. in phycology, was hired to assist Westman with laboratory work. Gauch soon learned computer programming, which he took to with alacrity. The Mendocino vegetation along an edaphic gradient, which was so clear and dramatic in the field, was proving hard to reproduce in ordinations using Bray-Curtis polar ordination, factor analysis or principal components analysis (PCA). After the Mendocino project was completed Whittaker encouraged Gauch to use his mathematical and computer programming talents to

103

continue testing ordination techniques (1972a, b, 1973i, 1977a, 1978d).

Whittaker recognized in his later papers on gradient analysis the potential usefulness of ordination for revealing the gradient structure of vegetation. Yet, he was concerned lest his work on ordination further alienate European ecologists of the classification tradition, whose friendship he had begun to court. In 1970 he delivered a paper (1972c) to the Rinteln symposium entitled 'Convergences of ordination and classification' in which he took pains to reassure European phytosociologists that ordination techniques, with their characterization of vegetation structure as continua, were not anti- thetical to the strict classificatory approaches of the Braun-Blanquet school. The paper was both an effort in international diplomacy and in theoretical ecology. Whittaker's standing among European ecologists was solidified, and in 1973 he assumed American editorship of the originally European journal, Vegetatio, a post which he held until his death. Soon after joining the editorial board, Whittaker helped formulate a new editorial policy for the journal emphasizing methodology and the conceptual unity of vegetation science in preference to descriptive papers of primarily regional interest. With the encouragement of Eddy van der Maarel, he siphoned a number of American papers on ordination and gradient analysis into the journal, where they appeared interleaved with European phytosociologlcal ones. From Wh~ttaker's perspective, he was seeking now to reunite the camps which had been set asunder by the cont inuum/communi ty controversy he had helped to flare in the 1950's. He edited and contributed extensively to a book on classification and ordination (1973g, 1978b, c) in which he tried to present with painstaking fairness the values of the various national traditions. These efforts reflected Whittaker's role as conciliator, and increasingly, as respected world authority.

In his years at Cornell, his career reached full bloom, and he embarked on an array of research projects. He continued his work on primary production, working up data for the Hubbard Brook forest (1974c) and the Santa Catalina Mountains (1975k) which he had obtained in the mid-1960's, and compiling summary tables of world primary production in a variety of ecosystems (1970c, 19731, 1975a, b). The work on primary production permitted calculation of nutrient budgets, and Whit-

104

taker conducted studies on nutrient stocks for the Brookhaven forest (19751), and Hubbard Brook forest (1979i). Whittaker also pursued his early interests in species diversity, stimulated in part by the attention given this topic by G. E. Hutchinson, R. H. MacArthur and their students. General patterns of plant species diversity did not emerge (1977g), and in his later studies of diversity in Mediterranean-climate ecosystems he sought not global generalizations, but local patterns which could be interpreted in light of peculiarities of site history and environment (e.g. 1979a, k, 1980, 1981b).

Although Whittaker was concerned about the environmental problems to which attention was being increasingly drawn in this period, he did not feel comfortable wandering from the scope of his specialization in natural community studies, and his written record on environment and conserva- tion issues is limited to discussions in the final pages of his text (1970c, 1975c), his contributions to the global carbon debate (1973m, 1978f), his 'Coda ' to the Brookhaven symposium on diversity and stability (1969c), and short comments on ecological effects of radiation, pesticides (1966b) and weather

R.H.Whit taker Total in

SCl • 10 II

g 5 0 0 '

. i ° a

u

o b

"- s

z

7 4 0 0 '

3 0 0 '

2 0 0 ,

• I 0 0

3 0

modification (1967a), and on the need to preserve natural diversity (e.g. 1960b, 19791-). Whittaker had served on the Governing Board of the Nature Conservancy while at Brooklyn College (1957-62), but resisted temptations to spread his energies in these directions in later years. He clearly preferred to concentrate on his own vegetation studies, and assist others in their efforts in this field.

The growth in Whittaker's reputation from 1965 to the time of his death was exponential_ One measure of this is the number of times articles of which he was senior author were cited in the scientific literature. Figure 1 shows the exponential rise in citations, to an average of over 400 per year at the time of his death. During this same period the number of citations of all scientific works also grew exponentially (Fig. 1), whereas the average number of citations per author remained constant at 6 to 8 per year. These trends can be interpreted as indi- cating that Whittaker had obtained 'authority' status during this period, and was being cited more commonly than other authors in part for this

Table 1. The ten works most often cited between 1965 and 1979, of which R. H. Whit taker was senior author. (Source of data: Science Citation Index).

• / 268

• 251 S C l t

w l )

199

i f 174

149

131

• 112

105 . . ~ * / : ( / / A v g - no . o f c i t a t i o n s

/ per author cited _ _ l £ . . . . . . . . . - - .

¢~s " =e '70 =9'1s 19"79 84

Fig. 1. The number of citations to works of which R. H. Whit taker (RHW) was senior author, the total number of citations for all authors (SCI), and the average number of citations per author cited, during the period 1965-1979.

No. of citations Communities and Ecosystem

78

R. H. Whittaker. 1st and 2nd editions. MacMillan: New York. 1970, 1975. R. H. Whittaker and P. P. Feeny. Allelochemics: chemical interactions between species. Science 171: 757-770. 1971. R. H. Whittaker. Gradie,at analysis of vegetation. Biol. Rev. 42:207 264. 1967. R. H. Whittaker. Dominance and diversity in land plant communities. Science 147: 250-260. 1965. R. H. Whittaker. Vegetation of the Great Smoky Mountains. Ecol. Monogr. 26:1 80. 1956. R. H_ Whittaker. Vegetation of the Siskiyou Moun- tains, Oregon & California. Ecol. Monogr. 30: 279-338. 1960. R. H. Whittaker. Evolution and measurement of species diversity. Taxon 21: 213-251. 1972. R. H. Whittaker. New concepts of kingdoms of organisms. Science 163:150 160. 1969. R H. Whittaker, ed. Handbook of Vegetation Scien- ce, Part V. Ordination and Classification of Vegeta- tion. Junk, The Hague. 737 p. 1973. R. H. Whittaker, Forest dimensions and production in the Great Smoky Mountains. Ecology 47: 103-121. 1966.

reason. The fact that his textbook was the most cited work lends credence to this interpretation, as the text was usually cited to document a fact or method, rather than to challenge one. Whittaker's ten most oft-cited works are listed in Table 1.

Whittaker received many inquiries from potential graduate students who sought his guidance, but he limited his acceptance of Ph.D. students for direct supervision to about one per year. During his lifetime, eight students completed their dissertations with him and four additional students completed a major portion of their dissertation work with him (Table 2). Whittaker also hosted several visiting scholars during his Cornell years. Latzi Fresco, Immanuel Noy-Meir, Mike Austin and Mark Hill all worked with Whittaker and Gaugh on ordination problems. The Israeli ecologists Zev Naveh and Avi Shmida worked with Whittaker on sampling and interpretation of diversity. He also supported the professional efforts of a number of ecologists from the Third World, Eastern bloc countries and China. His opposition to what he perceived to be totalitarian governments impelled him to provide particular support for so- called 'dissident' scientists. 8

Whittaker 's years at Cornell were marked both by a significant expansion and solidification of his reputation, and by major events, both joyful and tragic, in his personal life. Probably the most significant event of recognition occurred in 1974, when he was elected to the U.S. National Academy of Sciences. Beyond the Mercer Award, his appointment to Cornell, and his election as Vice- President of the Ecological Society of America in 1971, this was Whittaker's first truly outstanding, tangible evidence of recognition by the scientific establishment. The election marked a stage in Whittaker 's career which helped him personally to feel more confident, and ultimately more expansive, in his view of his professional role. Other honors were soon to follow. In 1976, he was appointed to an endowed chair at Cornell, the Charles Alexander Professorship in Biology. During the 1970's he was elected to the American Academy of Arts and Sciences, became an Honor- ary Member of the British Ecological Society and the Swedish Phytogeographical Society, and was President of the American Society of Naturalists at the time of his death.

In 1974 Clara Whittaker contracted cancer. Her

105

struggle with the disease lasted three years, until she succumbed, at Christmas time, 1977. Although Clara's prolonged illness upset Whittaker greatly he was stoically silent and many of his students and colleagues were not aware of the events that were troubling him. He turned to his traditional values for support, and increased even further the intensity with which he pursued his work. In the period following Clara's death, Whittaker developed a close friendship with his doctoral student, Linda Olsvig. I n October 1979, the two were married. Linda later accompanied Whittaker into the field on visits to Israel and South Africa, and took an active interest in his research.

Four months after his second marriage, Whit- taker complained of hip pain. X-rays revealed cancer in hip and lungs. 'We were told he had nine months, ' writes Linda, 'maybe two years if he was lucky. He wasn't. Bob set himself to complete as much as possible. We had planned to go to Israel, Spain and China - a wonderful trip to complete field work for a variety of projects. We did go to Israel, with Bob on crutches. He conducted himself with great courage and determination, to begin work he would never see completed, but to do his best to see that others would. '2 His health failed in September and he died October 20, 1980. Shortly before his death, the Ecological Society of America honored him with its highest award, that of Emi- nent Ecologist.

Although R. H. Whittaker will be remembered by ecologists most for his immensely important research contributions, those who met him found his personality equally arresting. As a person Whittaker was intense and reserved, with.an underlying warmth that sparkled intermittently at the surface. His intensity could be intimidating initially, but he was in fact most accessible and supportive to students and colleagues who sought to know him. Those who became closer to him realized that the intensity he radiated was in fact part of a much larger matrix of internal driving forces. These internal tensions, fired by the de- mands and lessons of his youth, served to forge his formidable intellect with profound sources of physical energy to produce a prodigious scientific output. These same driving forces appear to have been a source of the tenacity with which he was prepared to argue for the values he deeply held. While the weft of these influences on his work must

106

be left to the historian to unravel, it is to the benefit of the scientific community that they helped to produce the masterworks that are his legacy.

Research contributions

Gradient analysis

One of Whittaker's abiding research themes was the examination of patterns of biotic variation along environmental gradients. The me!hodology of 'gradient analysis' was developed during the initial Smoky Mountains studies and refined in the Siskiyou vegetation study. In both cases Whittaker traversed separate elevation and topographic- moisture gradients, sampling the Vegetation at each new exposure, elevation band, or conspicuous vegetation change. The sampling methods used in the Smoky Mountain study were crude and in- completely described, but for the Siskiyou study Whittaker settled on the 0.1 ha sampling units which became standard for his subsequent studies. This standardization allowed Whittaker and others to compare vegetation in geographically distinct areas on a scale hitherto impossible in North America.

Because site-to-site variation in species importance is notoriously large, Whittaker chose to smooth his species/site data by use of weighted averages, permitting underlying patterns to be more readily visible. He juxtaposed the smoothed elevation and topographic-moisture gradients as orthogonal axes, and indicated the location of physiognomic types within the axis space to form community mosaic diagrams. Borrowing from his wartime experience using weather maps with superimposed isobars, l~ he plotted species abundances on the mosaic as isopleths of constant average importance values of a species.

Whittaker's gradient analyses drew from un- usually large field data bases. He considered about 300 0.1-ha samples to be an appropriate minimum sample size for a vegetation study of a mountain range. ~ In this way, his work helped establish new standards for sample intensity in ecological field studies. Whittaker used his gradient studies to explore theoretical questions of vegetation structure, and to provide a framework for studies of ecosystem function. In the Great Smokies, the

gradient analysis initially served to test Clement- sian hypotheses of community structure, later was used as a basis for production studies (1966a), and s till later for studies of understory structure, pursued by Whittaker's graduate student, Susan Bratton.

Whittaker's gradient studies in the Siskiyou Mountains (1960b) appear to have been under- taken for at least three reasons. Firstly, these and the Klamath Mountains were a poorly known but floristically rich 'center' for Arcto-Tertiary forest elements. Elements of this flora also occurred in the Great Smoky Mountains (196 lc), providing a basis for crosscontinental comparison. Secondly, the variety of substrates in the Siskiyous provided a new environmental axis for testing gradient structure, and, in particular, to test Clements' notion that different substrates would not support different stable climax communities in the long term (1954c). Thirdly, the area was relatively close to Pullman, where Whittaker worked at the time of these field studies (1948-51). He later encouraged a graduate student, Mark Wilson, to study small scale patterns in serpentine vegetation of the Siski- yous.

Whittaker undertook a third gradient analysis study in the Santa Catalina Mountains of Arizona with William Niering (1964, 1965d, 1968b, c, d, 1975k). The Santa Catalina Mountains, 400 km southeast of the range on which Merriam's (1890) life-zone classification of communities was based, and site of a zonation study by Shreve (1915), provided another opportunity to test alternative hypotheses of species distributions along gradients. Whittaker and Niering found that both species and growth forms tended to vary continuously in abundance across major dominance types, thus suppor- ting Gleason's concepts. Whittaker and Niering further compared vegetation on the north and south sides of the mountain, documented differen- Ces between vegetation on limestone and granitic soils, and observed changes with elevation in life- forms, growthforms, geographic affinities, production, biomass and diversity.

In his final gradient studies, Whittaker sought to examine whether functional attributes varied along environmental gradients. He chose to observe changes in nutrient cycling patterns with changing soil attributes on a single substrate. One such study was located in the San Jacinto and Santa Rosa Mountains of southern California. Here vegetation

change could be observed in relation to soil properties along an elevation gradient from salt desert to subalpine forest. The bulk of the study was never prepared for publication. A mosaic diagram constructed by Whittaker and Niering did appear (Hanawalt & Whittaker, 1976b), but only the survey of changes in soil characteristics conducted in collaboration with R. B. Hanawalt was published in detail (Hanawalt & Whittaker, 1976b, 1977b, c). An additional study across an edaphic gradient from pygmy conifer scrub to giant redwood forests in Mendocino County had a similar theme (West- man & Whittaker, 1975b; Westman, 1975, 1978). Several graduate students associated with Whit- taker used gradient analysis as part of their disser~ tation research. Westman (1971, 1975) used these approaches in Mendocino; Peet in the Colorado Front Range (1975, 1978a, b, 1981); Wentworth (1976, 1981) for the Mule Mountains of southeastern Arizona; Lewin (1973, 1975) in the ravines of the Finger Lake Region of New York; and Olsvig in the pine barrens of Long Island (Olsvig et al., 1979b, 1980).

Whittaker also extended gradient analysis methods to applied ecological concerns. In collaboration with George Woodwell (1968f, 1973n) he examined a stress gradient generated by a gamma-emitting cesium source, and discussed Clementsian notions of retrogression in the context of continua. As part of his dissertation research under Whittaker's supervision, Steven Kessell (1979) developed an interactive computer program for forest fire management based on gradient models of vegetation and fuel load observed in Glacier National Park_

Although most of Whittaker's published research was in plant ecology, he retained a lifelong interest in animal ecology. While employed in the Zoology Department at Washington State University he conducted, with C_ W. Fairbanks, the study of copepods in small ponds (1958a) and wrote the monograph on Smoky Mountain foliage insects (1952). He later advised two students working on animal communities. Owen Sholes (1979) studied the response of spider communities to the phenol- ogical gradient imposed by the Solidago inflorescences they inhabitated. Steven Sabo (1980) applied gradient analysis methods to the community structure of subalpine birds of the White Mountains of New Hampshire.

107

The techniques of gradient analysis Whittaker developed provided a battery of methods for the study of vegetation as continua which represented an alternative to the methods of community-orien- ted phytosociologists. In so doing, Whittaker also developed the means to test hypotheses of gradient structure, and ultimately to replace the Clement- sian paradigm with a Gleasonian one.

Classification of communities and organisms

While Whittaker's conceptual style was generally to replace classifications with continua, he clearly recognized the need for categorization as a means to organize information and show relationships. Whittaker further recognized that the tremendous diversity of schemes for classifying vegetation was impeding communication between ecologists in various parts of the world. In 1961 he undertook to review community classification literature. The task was awesome as it required surveying much of the world literature of descriptive ecology for the preceding half century. A full year was spent on this project and the result was a book-sized review (1962a). The monograph won Whittaker the im- mediate respect of the European community of ecologists, who had not previously encountered a North American who understood so fully what they were trying to accomplish or how their methods related to those of other workers. The essential elements of the review were later presented as a chapter in Whittaker's book Ordination and Classi- fication of Communities (1973g, 1978b, c).

Whittaker also had a keen interest in the classification of kingdoms of organisms. He viewed the classical plant/animal dichotomy as highly artifi- cial. Applying ecological criteria, he recognized three groups with different modes of nutrition: primary producers or photosynthetic organisms (plants), consumers (animals), and decomposers (fungi and bacteria) which absorb rather than produce or consume their energy resources (1957b). This nutritional trichotomy provided the basis for his five-kingdom classification (1969d), the pro- caryotic and eucaryotic single-celled organisms composing the other two kingdoms. His five- kingdom proposal was widely adopted and served to establish Whittaker's reputation among systematists. Whittaker further expressed his interest in the evolution of organisms through his service as

108

Associate Editor of Paleobiology from 1967 to 1979.

Ordination and numerical classification

Whittaker's interest in methods of ordination derived from several sources, as described in his review of gradient analysis (1967b). He was aware of the early efforts by Curtis, Cottam, Mclntosh and Bray at the University of Wisconsin to rank samples of Wisconsin vegetation along axes of floristic similarity. The 'leading dominants' technique which Curtis & Mclntosh ( 1951) first attempted consisted of arranging stands to maximize the unimodality of distribution of dominant species. 'Adaptation values,' or weights corresponding to species peaks, were used to calculate weighted stand scores. Since the ultimate arrangement of stands derived from trial arrangement of stands based on their species composition, Whittaker labeled the procedure 'indirect gradient analysis.' In so doing he wished to emphasize that any environmental interpretation of the floristic axis of variation was derived strictly from knowledge of or assumptions about the environmental preferences of the flora, rather than from direct measurement of the habitats in which the plants were growing. Subsequent, more quantitative techniques of ordination developed by Bray & Curtis (1957) and others were included under the category 'indirect gradient analysis' because of their exclusive use of floristic information in axis derivation and interpretation.

Additionally, Whittaker was aware of the potential application to vegetation of the statistical tools of factor analysis and principal components (PCA) advocated by Goodall (1954), Dagnelie (1962) and Orl6ci (1966). None of these authors, including Whittaker, recognized at the time both that factor analysis assumed linear responses of species along environmental gradients, and that such linear responses would obscure the intrinsically nonlinear species distributions along environmental grar dients. The importance of these limitations was only first publicized by Swan in 1970.

Whittaker was drawn to use and evaluate ordination techniques because he recognized their potential for identifying gradients in species composition and their relationships to the underlying environment. Existing methods of community classification, be they taxonomic (Braun-Blauquet) or nu-

merical (e.g. Williams & Lambert, 1959), did not have this capacity. One of his first efforts in application of ordination methods was to encourage West- man to use factor analysis and polar(Bray-Curtis) ordination on the edaphic gradient from pygmy forest to redwoods in Mendocino County, Califor- nia. When factor analysis, PCA, polar ordination and various modifications of these failed to reveal the dramatic gradient visible in the field, Westman (1971, 1975) analyzed the statistical assumptions in factor analysis and PCA and noted problems in their application to vegetation data_ Whittaker subsequently encouraged Hugh Gauch, Jr., who had been working as a research assistant with him, to construct simulated data based on the notion that species should exhibit the same Gaussian response curves he had obtained with much data averaging and curve-smoothing from his Great Smoky Mountains study (Gauch & Whittaker, 1972a). Gauch & Whittaker (1972b, Whittaker & Gauch, 1973i) proceeded to analyze the ability of polar ordination and PCA to reproduce simulated species gradients or coenoclines of varying species richness (~-diversity) and gradient length (fl-diversity) as well as noise levels. This work helped to clarify the distortions which result from use of PCA and factor analysis discussed by Swan (1970), Noy- M eir & Austin (1971) and Beals (1973). Meanwhile, Whittaker edited and contributed extensively to a review of classification and ordination methods in community ecology (1973g).

Unreliability of PCA as a method for identification of important environmental gradients left only polar ordination as a commonly used method_ The shortcomings of polar ordination, including subjec- tivity in endpoint selection, were well known. Whittaker recognized that a logical step in the development of ordination methods for use in ecology would be development of techniques which explicitly assume non-linear species responses. He conceived of an ordination method in which species arrangement would be based on the best simultaneous fit to Gaussian curves. Gauch and Chase developed the algorithm, resulting in Gaussian ordination (Gauch, Chase & Whittaker, 1974a). Because species distributions are rarely perfectly Gaussian, even along strong environmental axes, and because single environmental gradients are rarely strong enough in nature to muffle the influence of secondary habitat factors, Gaussian

ordination proved only moderately effective in reproducing accurate floristic axes from either field data or simulated data.

Whittaker was aware of the limitations of the Gaussian assumption, and continued to seek more effective ordination methods (e.g., Kessell & Whit- taker, 1976c). He collaborated with other workers (I. Noy-Meir, M. Austin, and later M. O. Hill) who were seeking alternative approaches_ After Hill (1973) introduced reciprocal averaging, while poin- ting out its similarity with Benz6cri's (1963 1964) l'analyse factorielle des correspondances, Gauch, Whittaker & Wentworth (1977a) tested it with simulated coenoclines and field data. They found reciprocal averaging far superior in recovering known compositional gradients than any of the methods previously tested, but they also recognized several shortcomings they were unable to resolve. Whittaker invited Hill to Cornell in 1978-79 to work on improving the performance of reciprocal averaging. Hill developed DECORANA (detrended correspondence analysis) during this period (Hill & Gauch, 1980). The final ordination study in which Whittaker participated (Gauch, Whittaker & Singer, 198 la) compared several nonmetric multi- dimensional scaling techniques to reciprocal averaging and DECORANA_ DECORANA proved able to handle data sets of one and several major axes of variation better than the other techniques.

Whittaker's role throughout the development of ordination techniques was to encourage those with whom he worked (especially Hugh Gauch) to explore particular issues arising from his own gradient theories and field data, to contribute data for testing, to review the growing literature, to relate the mathematical results to theories of vegetation structure, and to help write and edit manuscripts on the subject. Whittaker's own limitations in mathematical and computer training inhibited him from becoming directly involved in the details of the programming, but his role as theoretician, critic and supporter of this work was crucial to its progress. With the development of DECORANA and its subsequent successful testing on simulated and field data, and with the production of the Cornell Ecology Program Series and the publication of the second edition of his ordination volume (1978c), Whittaker felt as though his involvement with the development of ordination techniques had reached a natural stopping point. 2

109

A second product of Mark Hill's stay at Cornel1 was the development of TWINSPAN, a new poly- thetic divisive classification algorithm and program which was a logical outgrowth of Hill's earlier indicator species analysis (Hill, Bunce & Shaw, 1975). The apparent superiority of this technique motivated Whittaker and Gauch to apply the techniques they had developed for evaluation of ordination methods to the analysis of numerical methods of classification.. Whittaker's final paper (Gauch & Whittaker, 1981d) in the area of multivariate analysis examined the performance of five hierarchical classification methods in relation to simulated and field data. With the clarity they had brought to earlier analyses, Gauch and Whittaker here carefully established criteria for evaluating classification performance, related the performance criteria to readily-conceptualized features of vegetation structure, and evaluated the performance of a range of classification methods.

The achievements of Gauch, Whittaker, Hill and coworkers in providing community ecologists with numerical methods of analysis of known performance and behavior will undoubtedly stand as signal contributions to the field of community ecology for many years.

Species diversity

Whittaker's interest in species diversity pre-dated the surge of attention given this topic from the late 1950's onward. In his early studies of the Smoky Mountains, Whittaker used his gradient model of vegetation as a framework for describing and interpreting the diversity of both trees (1956a) and foliage insects (I 952). Because of sample size variation he selected the index alpha of Fisher, Corbet & Williams (1943) to compare diversities. In the Siskiyou study, Whittaker (1960b) similarly included figures showing variation in diversity along the moisture and elevation gradients, and commented on the relation of richness to substrate and continentality.

Whittaker (1960b) studied diversity at several different scales. The diversity of a local community was termed alpha diversity in reference to the commonly used index of Fisher et al. As composition of a community at one point on a gradient does not reveal the rate of turnover of species composition along a gradient, he developed measures of this

110

gradient or coenocline differentiation, which he called beta diversity. He termed richness at the regional level gamma diversity and noted that numbers and rates of change in species can be conceptualizeff along a continuum of scales in the landscape (Whittaker, 1977g).

In 1965(c) Whittaker reviewed diversity measures and models of species abundance using a graph in which the logarithms of importance values were plotted sequentially from the most to the least important species. Whittaker suggested generalizations about the habitat conditions in which different characteristic shapes of these 'dominance- diversity' curves might he found. His discussion helped to show that Preston's lognormal model and MacArthur's broken-stick model of resource parti- tioning were only two of a wide range of dominance-diversity curves which might occur. The graphical format Whittaker introduced has subsequently been widely used in ecology, archae- ology, and other disciplines.

Following his studies of plant species diversity in the Smoky Mountains, Siskiyou's and Santa Ca- talina's, Whittaker sought the kind of generalizations MacArthur had so successfully described for birds. MacArthur had been able to predict bird species diversity from habitat characteristics, and with E. O. Wilson (1967), had developed a model for predicting species richness at equilibrium in isolated habitats based on simple indices such as habitat patch size and distance from a source for migrants. Such generalizations proved elusive for plants and Whittaker found only numerous, more qualified relationships specific to the taxon, community and area. Whittaker encouraged David Lewin (1975, Glenn-Lewin, 1977) to combine the mountain vegetation data sets with Lewin's data on ravines of upstate New York as part of a massive statistical exercise designed to reveal correlations with species diversity. Few significant correlations were obtained. Whittaker concluded his diversity review in 1977(g) with the observation that 'Al- though the study of diversity once sought the forms of exact science, it has found instead divergent relationships in different groups and areas, subject much less to prediction than to observation and evolutionary interpretation.' (p. 55).

After failing to find major unifying patterns in species diversity, Whittaker became increasingly interested in understanding what factors were re-

sponsible for the local variability in diversity patterns. He was intrigued by the idea of comparing isolated ecosystems which had developed under similar climatic regimes. The divergent properties of the lizard guilds Pianka studied in California, South Africa and Australia deeply impressed Whit- taker.11 Upon visiting an overgrazed pasture on Mt. Hermon in Israel, he was even more impressed to find a species richness per 0.1 ha almost double the highest he had previously recorded. Literature reports of very high levels of diversity and en- demism in the fynbos of South Africa and kwongan of south west Australia contrasted with the low diversity reported for chapparal and matto- ral. Intrigued by these observations Whittaker traveled extensively in California, Israel, southern Africa and Australia, comparing diversity in vegetation from regions with Mediterranean climates. 2 Unlike workers who were studying convergences between such regions, Whittaker focused on the causes of the differences in diversity. To him these differences appeared related to community age, substrate fertility and grazing stress. Several papers were published based on this work (Naveh & Whittaker, 1979a, j, 1980; Shmida & Whittaker, 1981b; Whittaker, Niering & Crisp, 1979k; Whit- taker, 1977g), but Whittaker's health failed him before he could undertake the planned major synthesis paper.

Community organization and dynamics

Whittaker's contributions to the theory of community structure and function followed several lines, most of which were related to his gradient view of vegetation organization. Although much of this work is interrelated conceptually, it is treated here under several headings.

Niche theoo'

Whittaker's conceptualization of niche differentiation along environmental gradients (1967b, 1969c, 1970c, 1975c), was his way of explaining the spread of modal peaks of importance value curves for dominant species along environmental gradients. On the one hand, niche differentiation served as a mechanism to explain the coenoclines he was observing along topographic moisture gradients on mountain slopes. On the other hand, the field data provided evidence for the niche hyper-

space model presented by G. E. Hutchinson in 1957. Thus Whittaker's concept of a n n - d i m e n - sional space of environmental factors in which Gaussian hypervolumes of species distributions could be found (1967b) corresponded well with Hutchinson's and MacArthur 's views of niche.

Because Whittaker 's view of niche structure developed from field data and his gradient concept of vegetation variation, he saw clearly the distinc- tion between species distributions along habitat factor axes, and species distributions within a stand in response to the same environmental complex_ The second was a concept of niche his gradient studies had not directly characterized, one much closer to Elton's original definition of niche as a species' role in the community. A new term was needed to embrace the concept that the species' environmental tolerances acted both within a site in defining the species niche volume, and between sites along major physical gradients of habitat. With Levin & Root (1973j), he suggested the term 'ecotope' for this combined niche-habitat concept. Whittaker then edited, with Levin, a book of historically important p a p e r s on niche theory (1975g).

Small-scale pattern analysis

Unlike R. H. MacArthur, whose inclinations led him further into mathematical conceptualizations of the niche, Whittaker 's field orientation led him towards attempts to identify the niche preferences of plants in the landscape. He encouraged his graduate student, Susan Bratton, to identify ways in which understory vegetation in the Great Smoky Mountains might be responding to environmental microheterogeneities. Whittaker recognized that MacArthur and other members of the Princeton school were similarly concerned with environmental heterogeneity or 'grain' in their theoretical studies of species diversity. He sought to make his own contributions at a theoretical level, and collaborated with S. A. Levin (1977h) and D. Good- man (1979h and manuscript) to try to relate population fluctuations due to climate and disturbance at the microscale to the maintenance of alpha diversity, and at a larger scale to beta and gamma diversities. His thinking on these subjects was still in an early stage of formulat ion at the time of his death, but the larger goal was already clear: to

111

relate individual demographic strategies and microsite heterogeneities to patterns of species diversity at the micro- and ultimately macro-scales.

Concurrent with these efforts, Whittaker began to obtain species richness data using nested qua- drats ranging f rom his initial 0.1 ha (1000 m 2)

samples through 100, 10, 5 down to 1 m 2 areas in vegetation from a variety of areas (Naveh & Whittaker, 1979a, j). These data could be used not only to construct species-area curves, but to observe floristic patterns at a range of scales. Bor- rowing f rom his ordination work, Whittaker sought to analyze micropatterns by ordinating his 1 m 2 samples using reciprocal averaging (with Naveh, 1979j, 1980; with Niering & Crisp, 1979k; with Gilbert & Connell, 1979g; with Shmida 198 lb; with Morris & Goodman, manuscript). The resulting data could be analyzed to detect microcommunity or synusual groupings, particularly for herb species, He had thus succeeded in characterizing 'patches' in the field, and was in a position to at tempt to relate these to theories of population dynamics, disturbance and habitat graininess, and in turn to diversity. This work of synthesis is, of course, the most difficult step, but Whittaker had brought his research to the brink of this phase at the time of his death.

Succession and climax pat tern

Whittaker's first publication (1951) was a challenge to the Clementsian notions of a regional monocl imax determined by climate, and to the notion of the seral replacement of one plant association by another. In this paper he suggested that by adopting Gleason's individualistic hypothesis of species distributions, the principle of community continuity (both in space and time) followed. Therefore, individual species, rather than whole communities, must replace each other over time. Site differences will induce differences in floristic composit ion from place to place, resulting not in a single climax type, or even a polyclimax, but in gradients of steady-state vegetation, a regional 'climax pattern. ' The regional climax pattern concept was a direct challenge to the Clementsian paradigm of the day, and set the stage for later analyses of succession on a species-by-species replacement basis pursued by Drury & Nisbet (1973), Connell & Slatyer (1977), Pickett (1976), Horn

112

(1975) and others in the 1970s. Whittaker himself returned to his concern with

successional theory in the 1970s (1974b, 1975d), and in his paper with Levin (1977h), sought to use microsite heterogeneities and disturbance to under- stand mosaic successions. He also encouraged two graduate students to study mosaic succession of forest trees. Runkle (1980), a student of Levin, examined tree fall replacement dynamics in the Smoky Mountains and Woods (1979, Woods & Whittaker, 1981c), studied cyclic replacement in northern hardwood forests. Although Whittaker (1974b, 1977h) recognized such processes as direct and cyclic succession and even established 'catego- ries' of successional types, he was not typological in outlook. Instead he viewed successional changes in time and space in relation to a number of axes, and constructed verbally a picture of forces leading toward vegetational stability, countered by forces of disturbance, leading to successional mosaics in the landscape. It was part of Whittaker's writing skill, as well as conceptual talent, to weave together threads (linear axes) representing a variety of environmental and biotic forces into a tapestry which, though complex in design, could evoke almost poetically a picture of the dynamic ecological processes in the mind of the reader.

Biomass, production and nutrient cycling

The earliest approaches to ecosystem function in Whittaker's research were his radiophosphorus tracer studies in aquaria, performed at Hanford Laboratories between 1952-54 (1961b). This early study is remarkable for its effective characterization of phosphorus movement using the compart- ment model pioneered by E. P. and H. T. Odum only a few years earlier. The aquarium study applied several concepts which were later to find widespread use in nutrient cycling studies: the concentration ratio for elements between trophic levels; turnover rates and decay constants; and the determination of transfer rate constants between compartments.

Whittaker's first major attempt to characterize production in terrestrial ecosystems was his study of aboveground production along the elevational gradient in the Great Smoky Mountains. The novelty of his first approach to production estimation was his extension of production measurement

to all aboveground parts of the vegetation: flowers, fruits, branch wood and bark, and stem wood and bark, of trees and shrubs_ He borrowed the concept of volumetric measurement of boles from foresters, but developed his own system for measuring branch biomass and production using both annual xylem ring widths and bud scale scars. His development of logarithmic regressions based on allometric growth assumptions permitted the estimation of forest biomass and production in forests of similar species and physiognomy, but on different sites. With Violet Garfine, a Master's student, he also explored the use of chlorophyll content as an index of productivity (1962c).

In collaboration with G. M. Woodwell, Whit- taker expanded and refined his production measurement methods into the technique dubbed 'dimension analysis.' With Woodwell, he developed measures of surface area of forests, which were the first such attempts. During field work in the 1960's, Whittaker applied the dimension analysis approach to vegetation gradients in the Santa Catalina Mountains of Arizona (1975k), the Hubbard Brook watershed in New Hampshire (1974c) and the San Jacinto Mountains of California, although the latter data remain largely unpublished. His discussion of the Santa Catalina data (1975k) was particularly insightful in tracing patterns of primary production and biomass in relation to environmental gradients. Westman and Whittaker (1975b) also applied the technique to Californian coastal conifer forests. Whittaker did not confine use of the method to production measurement, but also began to use dimension analysis techniques, along with tissue nutrient analyses, to develop cation budgets, first for the Brookhaven forest (19751), and later for the Hubbard Brook forest (1979i).

Whittaker was able to build an extensive base for comparison of biomass and production characteristics in plant communities from his own data. Publication of the massive data tables resulting from the studies was often difficult because of journal policies, but Whittaker argued convincing- ly that future studies of production depended on the availability of species-specific regression equations, and that recognition of global patterns in production depended on reliable data being available from many sites. Whittaker also compiled a table of primary production and biomass in world ecosystems, first in his text (1970c), later (1973m)

refining the estimates for aquatic communities, with G. E_ Likens. This table became a basis for major discussions of global primary production totals, and Whittaker & Lieth edited a volume on the subject in 1975(a). The global productivity data gave Whit taker an opportuni ty to contribute to the debate on vegetation as a net source or sink for atmospheric carbon dioxide (1975i and 19780.

Whittaker 's major contributions to production studies were firstly in his development of a detailed system for static production measurement in temperate forest ecosystems, and secondly in his com- pilation of extensive sets of data, both from his own work and f rom the world literature, to reveal major patterns of plant primary production in relation to environmental factors. His extension of this work to nutrient cycling produced an extensive and useful base of empirical information, although as he himself commented in the Hubbard Brook paper (1979i), the degree of effective generalization about patterns of nutrient cycling was somewhat dis- appointing.

Allelochemistry

The contributions of R. H. Whittaker to allelochemistry derive exclusively from his analysis and synthesis of existing literature on the subject. In 1968, C. H. Muller invited him to be the keynote speaker at a National Academy of Sciences symposium on plant-plant interactions. Muller provided Whittaker with a brief bibliography of recent publications on allelopathy. With characteristic thoroughness, Whittaker researched the literature and provided a lecture entitled 'The Chemistry of Communit ies ' (1971 f) which impressed many of the researchers assembled. 3 Whittaker had used both his skill at synthesis of concepts, and his powerful lecturing style, toes tabl ish himself in a field that was entirely new to him, and in which he had not done and never subsequently did any original research. It is somewhat ironic that the paper on allelochemics published with Paul Feenyin 197 l(a), which was an expansion of the themes in this lecture, became the single most frequently cited article Whittaker ever wrote (Table 1).

Whittaker developed a strong friendship with C. H. Muller while both were at campuses of the University of California in the mid-1960's. They shared a common training f rom Vestal at Illinois, 3

113

and a common cultural and political background from their youths, Muller having been raised in southern Texas. Whittaker funded a chemical analytic facility in Muller's laboratory in the late 1960's. In his review of allelopathy, Whittaker was a strong supporter of Muller's work, although its applicability to field situations was being challenged by others (e.g. Bartholomew, 1970). The ultimate predictive power of this work on allelopathy is still being evaluated. Whittaker 's support of it is therefore still a matter for scientific controversy. Whittaker 's contribution to the present has lain in his synthesis of evidence from a variety of areas of allelochemistry, and his postulation of a number of evolutionary mechanisms for the origin and selection of such phenomena.

Concluding remarks

R. H. Whittaker 's work was characterized on the one hand by a search for patterns and generalization about the structure of biological communities, and on the other by the consistent recognition of the resistance of nature to categorization, classification and simplification_ Thus Whittaker emphasized the continuity in the distribution of characters in nature, introducing a calculus to realms of ecological theory hitherto transfixed by categorization. Indeed, it was characteristic of Whittaker to ask of an existing categorical generalization in ecology, How can this generalization be conceptualized in terms of gradients? How can this axis of change be viewed as one of many axes in an n-dimensional space? This same pattern of thought characterized first Whittaker's approach to the distribution of species in relation to environmental factors in space (the gradient theory) and time (the climax pattern), and subsequently to communities and ecosystems.

Whittaker typically selected for study areas with little apparent theoretical cohesion. In both the thoroughness of his reviews and the eclecticism of his approach he was unequalled among ecologists of his time, save perhaps for G. E. Hutchinson. There was not always unanimity of acceptance of the newcomer's interpretations of the literature, and to some Whittaker seemed overly dogmatic once his views were presented. Nonetheless, Whit- taker's reviews were consistently thorough and insightful.

114

Whittaker recognized the importance of hypothesis testing but often preferred to reach generalizations by induction. Typically he would identify a problem, collect the appropriate data, and then interpret the results in what seemed to him the most intuitively satisfying manner. He often smoothed or transformed data to better fit an expected pattern, and deleted 'outliers.' While this ran the risk of overlooking the significant anomaly, it also succeeded in isolating broad patterns which bore well the test of 'replication' from other studies. While Whittaker was prone to draw 'smoothed' Gaussian curves through scatters of points with an impunity that amazed and alarmed his graduate students, he nevertheless resisted the temptation to become too entranced with elegance and simplicity. It was perhaps the latter side of him, along with his own limitations in mathematical training, that dissu- aded him from pursuing at greater lengths the kind of mathematical model building characteristic of MacArthur and his followers. A passage at the end of Whittaker's article (with S. Levin, 1977h) on mosaic phenomena concisely illustrates his viewpoint (p. 136):

Ecological theory is not precluded by, but should make realistic allowance for, the intrinsic diversity of ecological phenomena; and ecological research must often center more on analysis, interpretation, comparison, and modeling of cases than on widely applicable generalization. Ecologists have sought a theory or master plan of evolution permitting interpretation of communities through a limited number of strongly linked and widely significant relationships. Such a theory is naturally desired by ecologists as scientists; b u t . . , there may be no master plan except, perhaps, the evolution of such a diversity of relationships as to frustrate that desire.

It is difficult at this point in the history of the discipline to assess with certainty which of Whit- taker's contributions to the science of ecology will be most profound or longlasting. Nevertheless, certain hallmarks of his contributions are already clear: the challenge posed to classificatory approaches to vegetation structure led to the development of the set of theoretical and methodological developments now known as gradient analysis. Although he gave credit to Ramensky and Gleason

for the origins of this tradition and to Curtis and Mclntosh for inspiring indirect gradient analysis approaches, it was Whittaker who, more than anyone, solidified the theoretical, methodological and empirical bases for this approach. More than any other unifying notion, it was the conceptualization and demonstration of the continuity of species' response to environmental gradients that characterized Whittaker's work, and which seems likely to be his most enduring contribution.

Through his personal diplomacy, Whittaker built bridges between American and European ecologists over waters which he had troubled with his challenges to phytosociological theories and methods of classification. His reviews of classification and ordination studies, and his global studies of diversity and productivity helped inspire some North American ecologists to increase contacts and collaboration with ecologists beyond their borders.

Whittaker played a major role in generating and disseminating methodologies for the study of plant communities (diversity measurement, gradient analysis, ordination), which he furthered by encouraging Hugh Gauch, Jr. to write computer programs for widescale distribution (Cornelt Ecology Programs series). Further, Whittaker's authority status gave his methods a credibility that led to some degree of international standardization of methods. A particularly clear example of this function is the widespread use of the 0.1 hectare samples by workers measuring species richness in vegetation. The loss of this authority in a field short on internationally acknowledged authorities can have a major impact in loosening the cohesion that he helped foster.

Finally, many of Whittaker's contributions remain unpublished in the marginalia of countless manuscripts, and in the correspondence files of myriad colleagues. In addition to his formal editorial duties, Whittaker corresponded with and encouraged an immense number of students and colleagues, many of whom he never personally met. He also delivered a large number of lectures and seminars, characteristically with a sharpness of delivery and density of information that impressed even skeptical audiences. In this way, his influence spread far beyond his published work.

The breadth of Whittaker's influence is a reflection of his own enormous grasp of the fields of ecology and evolutionary biology. Whittaker

combined this immensity of vision with a passionate attention to detail and with a personal commitment to collect accurate and abundant supp orting data. In his analysis and synthesis of this material he used an intensity of written and oral delivery that reflected his own intellectual ferment. To those who knew him, and are yet to know him, through his writings and scientific influence, he is destined to remain one of the most important figures in community ecology to have lived in this century.

Acknowledgements

We thank the following for responding to our requests for information: B. Chabot, H. Gauch, Jr., D. C. Glenn-Lewin, M. O. Hill, S. C. Kendeigh, S. Levin, H. Lieth, E. van der Maarel, L. T. McHargue, C. H. Muller, W. A. Niering, L. Olsvig-Whittaker, R. Tobey, W. H. Schlesinger, O. Sholes, A. W. Smith, T. R. Wentworth, G. M. Woodwell. The following people provided comments o n earlier drafts of the article: E. van der Maarel, P. L. Marks, R . P . McIntosh, W.A. Niering, L. Olsvig-Whit- taker, W. H. Schlesinger, R. S. Westman.

Notes

The letters cited were addressed to the authors, and are on file in the History of Ecology Archives, University of Georgia, under 'R. H_ Whittaker'.

1. Alice Whittaker Smith, personal communication to W. E. Westman, September 29, 1981.

2. Letter; Linda Olsvig-Whittaker, December 3 l, 1980. 3. Letter; Cornelius H. Muller, January 12, 1981. 4. R. H, Whittaker (1972e). 5. Letter; S. C. Kendeigh, February 3, 1981. 6. Letter; David C. Glenn-Lewin, April 7, 1981. 7. R. H. Whittaker, personal communication to W. E.

Westman, 1969. 8. Letter; Simon Levin, March 18, 1981. 9. Letter; Owen Sholes, January 15, 1981.

10. Letter; Helrnut Lieth, February 23, 1981. 11. R. H, Whittaker, personal communication to R. K. Peet,

1972.

Appendix: Publications of Robert H. Whittaker

1948 Whittaker, R. H. A vegetation analysis of the Great Smoky Mountains. Dissertation, University of Illinois, Department of Zoology.

1951

1952

1953

1954a.

1956a.

b.

1957a.

1958a.

115

Whittaker, R. H. A criticism of the plant association and climatic climax concepts. Northwest Sci. 26: 17-31. Whittaker, R. H. A study of summer foliage insect communities in the Great Smoky Mountains. Ecol. Monogr. 22: 1-44. Whittaker, R. H. A consideration of climax theory: The climax as a population and pattern. Ecol. Monogr, 23: 41-78. Whittaker, R. H. Plant populations and the basis of plant indication. (German summ.) Angew. Pflanzensoziol. (Wien), Festschr. Aichinger 1: 183-206.

b. Whittaker, R. H. The ecology of serpentine soils. I. Introduction. Ecology 35: 258-259.

c. Whittaker, R. H. The ecology of serpentine soils. IV. The vegetational response to serpentine soils. Ecology 35:275 288. Whittaker, R. H. Vegetation of the Great Smoky Mountains. Ecol. Monogr. 26: 1-80. Whittaker, R. H. In honor of Erwin Aichin- ger. Review of Festschrift for Erwin Aichin- ger zum 60 Geburts tag. 1954. Ecology 37: 396-397.

c. Whittaker, R. H. A new Indian Ecological Journal. Review of Bulletin of the Indian Council of Ecological Research, Vol. 1. Ecology 37: 628. Whittaker, R. H. Recent evolution of ecological concepts in relation to the eastern forests of North America. Am. J. Bot. 44: 197-206. Reprinted in, Fifty Years of Bota- ny: Golden Jubilee Volume of the Botanical Society of America, W. C. Steere, ed., p. 340-358. McGraw-Hill, New York, 1958.

b. Whittaker, R. H. The Kingdoms of the living world. Ecology 38: 536-538.

c. Whittaker, R. H. Gradient analysis in agri- cultural ecology. Review of H. Ellenberg. 1950-54. Landwirtschaftliche Pflanzenso- ziologie. Ecology 38: 363-364.

d. Whittaker, R. H. Two ecological glossaries and a proposal on nomenclature. Ecology 38: 371. Whittaker, R. H. & C. W. Fairbanks. A study of plankton copepod communities in the Columbia Basin, southeastern Washing- ton. Ecology 39: 46-65. Reprinted in, Readings in Population and Community Ecology, W. E. Hazen, ed., p. 369-388.

116

Saunders, Philadelphia, 1964. b. Whittaker, R. H. A manual of phytoso-

ciology. Review of Bharucha, F. R_ and W. C. de Leeuw. 1957. A practical guide to plant sociology for foresters and agricul- turalists. Ecology 39: 182.

c. Whittaker, R. H. The Pergamon Institute and Russian journals of ecology. Ecology 39: 182-183. Whittaker, R. H. On the broad classification of organisms. Q. Rev. Biol. 34: 210-226_ Whittaker, R. H. Ecosystem. In McGraw- Hill Encyclopedia of Science andTechnol- ogy, p. 404-408. McGraw-Hill, New York.

b. Whittaker, R. H. Vegetation of the Siskiyou Mountains, Oregon and California. Ecol. Monogr. 30:279 338.

c. Whittaker, R. H. A vegetation bibliography for the northeastern states. Review of F. E. Egler. 1959. A cartographic guide to selected regional vegetation literature where plant communities have been described. Ecology 41 : 245-246, Whittaker, R. H. Estimation of net primary production of forest and shrub communities. Ecology 42: 177-180.

b. Whittaker, R. H. Experiments with radiophosphorus tracer in aquarium microcosms. Ecol. Monogr. 31: 157-188.

c. Whittaker, R. H. Vegetation history of the Pacific Coast states and the 'central' significance of the Klamath Region. Madrofio 16:5 23. Whittaker, R. H. New Serials. Ecology 42: 616. Whittaker, R. H. Classification of natural communities. Bot. Rev. 28:1 239. Reprin- ted by Arno Press, New York, 1977.

b. Whittaker, R. H. Net production relations of shrubs in the Great Smoky Mountains. Ecology 43: 357-377.

c. Whittaker, R. H_ & V. Garfine. Leaf characteristics and chlorophyll in relation to exposure and production in Rhododendron maximum. Ecology 43: 120-125.

d. Whittaker, R. H. The pine-oak woodland community. Review ofJ. T. Marshall. 1957. Birds of pine-oak woodland in southern Arizona and adjacent Mexico. Ecology 43: 180-181.

1959

1960a.

1961a.

d.

1962a.

1963a. Niering, W. A., R. H. Whittaker & C. H. Lowe. The saguaro: A population in relation to environment. Science 142(3588): 15-23.

b. Whittaker, R. H. Essays on enchanted is- lands. Review of The Enchanted Voyage and Other Studies, by G. E. Hutchinson. Ecology 44: 425.

c. Whittaker, R. H. Net production of heath balds and forest heaths in the Great Smoky Mountains. Ecology 44: 176-182.

d. Whittaker, R. H., N. Cohen & J. S. Olson. Net production relations of three tree species at Oak Ridge, Tennessee. Ecology 44: 806-810.

1964 Whittaker, R. H. & W. A. Niering. Vegeta- tion of the Santa Catalina Mountains, Ari- zona. 1. Ecological classification and distribution of species. J. Ariz. Acad. Sci. 3: 9-34.

1965a. Niering, W. A. & R. H. Whittaker. The saguaro problem and grazing in southwes- tern national monuments. Natl. Parks Mag. 39(213): 4-9.

b. Whittaker, R. H. Branch dimensions and estimation of branch production. Ecology 46: 365-370.

c. Whittaker, R. H. Dominance and diversity in land plant communities. Science 147 (3655): 250-260.

d. Whittaker, R. H. & W_ A. Niering Vegeta- tion of the Santa Catalina Mountains, Ari- zona: A gradient analysis of the south slope. Ecology 46: 429-452.

1966a. Whittaker, R. H. Forest dimensions and production in the Great Smoky Mountains. Ecology 47: 103-121.

b. Woodwell, G. M., W. M. Malcolm & R. H. Whittaker. A-bombs, bugbombs, and us. NAS-NRC Symposium on 'The Scientific Aspects of Pest Control', by the Brookhaven National Laboratory, U.S. Atomic Energy Commission.

1967a. Whittaker, R. H_ Ecological implications of weather modification. In Ground Level Climatology, R. H. Shaw, ed., p. 367-384. American Association for the Advancement of Science, Washington, Publ. 86.

b. Whittaker, R. H. Gradient analysis of vegetation. Biol. Rev. 42: 207-264.

c. Whittaker, R. H. & G_ M. Woodwell_ Sur-

face area relations of woody p!ants and forest communities. Am. J. Bot. 54:931-939.

d. Woodwell, G. M. & R. H. Whittaker. Pri- mary production and the cation budget of the Brookhaven forest. Symposium on Pri- mary Productivity and Mineral Cycling in Natural Ecosystems, H_ E_ Young, ed., p_ 151-166. University of Maine Press, Orono.

1968a. Frydman, I. & R. H. Whittaker. Forest associations of southeast Lublin Province, Poland. (German summ.) Ecology 49: 896-908.

b_ Whittaker, R. H., S. W. Buol, W. A. Niering & Y. H_ Havens. A soil and vegetation pattern in the Santa Catalina Mountains, Arizona. Soil Sci. 105: 440-450.

c. Whittaker, R. H. & W. A_ Niering. Vegeta- tion of the Santa Catalina Mountains, Ari- zona. III. Species distribution and floristic relations on the north slope. J. Ariz. Acad. Sci. 5: 3-21.

d. Whittaker, R. H_ & W. A. Niering. Vegeta- tion of the Santa Catalina Mountains, Ari- zona. IV. Limestone and acid soils. J. Ecol. 56: 523-544.

e. Whittaker, R. H. & G. M. Woodwell. Di- mension and production relations of trees and shrubs in the Brookhaven forest, New York. J. Ecol. 56: 1-25.

f. Woodwell, G. M. & R. H. Whittaker. Ef- fects of chronic gamma irradiation on plant communities. Q. Rev. Biol. 43:42 55.

g. Woodwell, G_ M. & R. H. Whittaker. Pri- mary production in terrestrial ecosystems. Am. Zool. 8 :19 30. Reprinted in Energy Flow and Ecological Systems, F. B. Turner, ed.

1969a. Whittaker, R. H. A view toward a National Institute of Ecology. Ecology 50: 169-170.

b. Whittaker, R. H. Een nieuwe indeling van de organismen. Natuur en Techniek 37: 124 132.

c. Whittaker, R. H. Evolution of diversity in plant communities. In Diversity and Stabil- ity in Ecological Systems, Brookhaven Symposia in Biology, No. 22, p. 178-195.

d. Whittaker, R. H. New concepts of kingdoms of organisms. Science 163(3863): 150 160.

e. Whittaker, R. H. & G. M_ Woodwell. Struc- ture, production, and diversity of the oak-

1970a.

1971a.

117

pine forest at Brookhaven, New York. J. Ecol. 57: 155-174. Bormann, F. H., T. G. Siccama, G. E. Likens & R. H. Whittaker. The Hubbard Brook ecosystem study: Composition and dynamics of the tree stratum. Ecol. Monogr. 40: 373-388_

b. Brown, W. L., Jr., T. Eisner & R. H. Whittaker. Allomones and kairomones: Transspecific chemical messengers. Bio- Science 20: 21-22.

c. Whittaker, R. H. Communities and Eco- systems. Macmillan, New York. xi + 162 pp. Reprinted in Japanese edition, Tokyo, 1974.

d. Whittaker, R. H. Neue Einteilung der Or- ganismenreiche. Umschau 16:514-515.

e_ Whittaker, R. H. Taxonomy. In McGraw- Hill Yearbook of Science and Technology 1970, p. 365 369. McGraw-Hill, New-York.

f. Whittaker, R_ H. The biochemical ecology of higher plants. In Chemical Ecology, E. Sondheimer and J. B. Simeone, eds., p. 43-70. Academic Press, New York.

g. Whittaker, R. H. The population structure of vegetation. In Gesellschaftsmorphologie (Strukturforschung), (German summ.), R. Ttixen, ed., p. 39-62. Ber. Symp. Int. Ver. Vegetationskunde, Rinteln, 1966. Junk, The Hague.

h. Woodwell, G. M. & R. H. Whittaker. Ioni- zing radiation and the structure and func- tions of forests. In Gesellschaftsmorphologie (Strukturforschung), (German summ.), R. Ttixen (ed.), p. 334-339_ Ber. Symp. Int. Ver. Vegetationskunde, Rinteln, 1966. Junk, The Hague. Whittaker, R. H. & P. P. Feeny. Allelo- chemics: Chemical interactions between species. Science 171 (3473)- 757 770.

b. Whittaker, R. H. & G. M. Woodwell. Evo- lution of natural communities. In Ecosys- tem Structure and Function, Proceedings of the 31st Annual Biology Colloquium, J. A. Wiens, ed., p. 137 156. Oregon State Uni- versity Press, Corvallis.

c. Whittaker, R. H. & G. M. Woodwell. Mea- surement of net primary production of forests. Reprinted in Productivity of Forest Ecosystems (French summ.), Proceedings of the Brussels Symposium, 1969, P. Duvig- neaud, ed., p. 159-175. Unesco, Paris.

118

d. Brussard, P. F., S. A. Levin, L. N. Miller & R. H. Whittaker. Redwoods: a population model debunked. Science 175: 435-436.

e. Whittaker, R. H. Dry weight, surface area, and other data for individuals of three tree species at Oak Ridge, Tennessee. In P.

Sollins & R. M. Anderson (eds.), Dry- weight and Other Data for Trees and Woody Shrubs of the Southeastern United States. Oak Ridge National Lab. Pub. ORNL-IBP- 71-6. p. 37-38.

f. Whittaker, R. H. The chemistry of communities. In Biochemical Interactions Among Plants. National Academy of Scien- ces. p. 10-18.

1972a. Gauch, H. G., Jr. & R. H. Whittaker. Coenocline simulation. Ecology 53:446-451.

b. Gauch, H. G., Jr. & R. H. Whittaker. Comparison of ordination techniques. Ecol- ogy 53: 868-875.

c. Whittaker, R. H. Convergences of ordination and classification. Reprinted In Basic Problems and Methods in Phytosociology (German summ.), Ber. Symp. Int. Ver. Vegetationskunde, Rinteln, 1970, R. Ttixen, ed., p. 39-55. Junk, The Hague.

d. Whittaker, R. H. Evolution and measurement of species diversity. Taxon 21 : 213-251.

e. Whittaker, R_ H. A hypothesis rejected: the natural distribution of vegetation. In W. A.

Jensen & F. B. Salisbury (eds.), 1972, Botany: An Ecological Approach. Wads- worth, Belmont, California, p. 689-691. Reprinted in W. A. Jensen et aL (eds.), 1979, Biology. Wadsworth, Belmont, California, p. 474-476.

1973a. Cottam, G., F. G. Goff& R. H. Whittaker. Wisconsin comparative ordination. In Ordi- nation and classification of communities, R. H. Whittaker, ed., p. 193-221. Junk, The Hague.

b. Whittaker, R. H. Approaches to classifying vegetation. In Handbook of Vegetation Science, Part V: Ordination and Classifica- tion of Vegetation, R. H. Whittaker, ed., p. 325 354. Junk, The Hague.

c. Whittaker, R. H. Community, biological. In Encyclopaedia Britannica, p. 1027 1035. 15th edition.

d. Whittaker, R. H. Direct gradient analysis:

Results. In Handbook of Vegetation Scien- ce, Part V: Ordination and Classification of Vegetation, R. H. Whittaker, ed., p. 35 51. Junk, The Hague.

e. Whittaker, R. H. Direct gradient analysis: Techniques. In Handbook of Vegetation Science, Part V: Ordination and Classifica- tion of Vegetation, R. H. Whittaker, ed., p. 9 31. Junk, The Hague.

f. Whittaker, R. H. Dominance-types. In

Handbook of Vegetation Science, Part V: Ordination and Classification of Vegetation, R. H_ Whittaker, ed., p. 389-402. Junk, The Hague.

g. Whittaker, R. H., ed. Handbook of Vegetation Science, Part V: Ordination and Classification of Vegetation. Junk, The Hague. 737 pp.

h. Whittaker, R_ H. Introduction. In Hand- book of Vegetation Science, Part V: Ordination and Classification of Vegetation, R. H. Whittaker, ed., p. l-6. Junk, The Hague_

i. Whittaker, R. H_ & H. G. Gauch, Jr. Evaluation of ordination techniqhes. In Handbook of Vegetation Science, Part V: Ordination and Classification of Vegetation, R. H. Whittaker, ed., p. 289-321. Junk, The Hague.

j. Whittaker, R. H., S. A. Levin & R. B. Root. Niche, habitat, and ecotope. Am. Nat. 107: 321-338.

k. Whittaker, R. H. & G. E. Likens. Carbon in the biota. In Carbon and the Biosphere, G. M. Woodwell & E. V. Pecan, eds., 281-302. U.S. Atomic Energy Commission, CONF- 720510, Springfield, Virginia.

1. Whittaker, R. H. & G. E. Likens. Introduction. In The Primary Production of the Biosphere, Syrup. given at the Second Congress of the American Institute of Biological Sciences, Miami, 1971. Human Ecol. 1: 301-302.

re.Whittaker, R. H. & G. E. Likens. Primary production: The biosphere and man. In The Primary Production of the Biosphere, Syrup. given at the Second Congress of the American Institute of Biological Sciences, Miami, 1971. Human Ecol. 1:357 369.

n. Whittaker, R. H. & G. M_ Woodwell.

1974a.

1975a.

Retrogression and coenocline distance. In

Handbook of Vegetation Science, Part V: Ordination and Classification of Vegetation, R. H. Whittaker, ed., p. 55-73. Junk, The Hague_ Gauch, H. G., Jr., G, B. Chase & R. H. Whittaker. Ordination of vegetation samples by Gaussian species distributions_ Ecology 55: 1382-1390.

b_ Whittaker, R. H. Climax concepts and recognition. In Handbook of Vegetation Science, Part VIII: Vegetation Dynamics, R. Knapp, ed., p. 139-154. Junk, The Hague.

c. Whittaker, R. H., R. H. Bormann, G. E. Likens & T_ G. Siccama. The Hubbard Brook ecosystem study: Forest biomass and production. Ecol. Monogr. 44: 233-254. Lieth, H. & R. H. Whittaker (eds.) The Primary Productivity of the Biosphere. Springer-Verlag, New York. 339 pp.

b_ Westman, W. E. & R. H. Whittaker. The pygmy forest region of northern California: Studies on biomass and primary productivity. J. Ecol. 62: 493-520.

c. Whittaker, R. H. Communities and Eco- systems. 2nd Edition. Macmillan, New York. 385 pp. Reprinted in Japanese edition, Tokyo, 1978.

d. Whittaker, R_ H. Functional aspects of succession in deciduous forests. In Sukzes- sionsforschung (German summ.), W. Schmidt, ed_, p. 377-405. Ber. Syrup. Int. Ver. Vegetationskunde, Rinteln, 1973.

e. Whittaker, R. H. The design and stability of plant communities. In Unifying Concepts in Ecology, W. H. van Dobben & R. H. Lowe- McConnell, eds., p. 169-181. Report of Plenary Sessions, 1st International Congress of Ecology, The Hague, 1974. Junk, The Hague & Pudoc, Wageningen.

f. Whittaker, R. H. Vegetation and parent material in the western United States. In

Vegetation und Substrat, (German summ.), H, Dierschke, ed., p. 443-465. Ber. Symp. Int_ Ver. Vegetationskunde, Rinteln, 1969.

g. Whittaker, R. H. & S, A_ Levin (eds_) Niche: Theory and Application_ Benchmark Papers in Ecology. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania_ 448 pp,

h. Whittaker, R. H., S. A. Levin & R. B. Root.

119

On the reasons for distinguishing 'niche, habitat, and ecotope.' Am. Nat. 109: 479-482.

i. Whittaker, R. H. & G. E. Likens. The biosphere and man. In Primary Productivity of the Biosphere, H. Lieth and R. H. Whittaker, eds., p. 305-328. Springer Verlag, New York.

j. Whittaker, R. H. & P. L. Marks. Methods of assessing terrestrial productivity. In Primary Productivity of the Biosphere, H. Lieth and R. H. Whittaker, eds., p_ 55-118. Springer Verlag, New York.

k. Whittaker, R. H. & W. A. Niering. Vegetation of the Santa Catalina Mountains, Arizona_ V. Biomass, production, and diversity along the elevation gradient. Ecology 56:771 790.

1. Woodwell, G. M., R. H. Whittaker & R. A. Houghton. Nutrient concentrations in plants in the Brookhaven oak-pine forest. Ecology 56:318 332.

1976a. Gauch, H. G., Jr. & R. H. Whittaker. Simulation of community patterns. Vege- tatio 33:13 16.

b. Hanawalt, R. B_ & R. H. Whittaker. Altitudinally coordinated patterns of soils and vegetation in the San Jacinto Moun- tains, California. Soil Sci. 121: 114-124.

c. Kessell, S. R. & R. H. Whittaker_ Comparisons of three ordination techniques. Vegetatio 32: 21-29.

1977a. Gauch, H. G,, Jr., R. H. Whittaker & T_ R. Wentworth, A. comparative study of reciprocal averaging and other ordination techniques. J. Ecol_ 65: 157-174.

b. Hanawalt, R. B. & R. H. Whittaker. Altitudinal patterns of Na, K, Ca and Mg in soils and plants in the San Jacinto Mountains, California. Soil Sci. 123:25 36.

c. Hanawalt, R. B. & R. H. Whittaker. Altitudinal gradients of nutrient supply to plant roots in mountain soils. Soil Sci. 123: 85-96.

d. Noy-Meir, I. & R. H. Whittaker. Continuous multivariate methods in community analysis: Some problems and developments. Vegetatio 33:79 98.

e. Whittaker, R. H. Animal effects on plant species diversity. In Vegetation und Fauna,

120

R. Ttixen, ed., p. 409-425. Ber. Symp. Int. Ver. Vegetationskunde, Rinteln, 1976. Cra- mer, Vaduz.

f. Whittaker, R_ H. Broad classification: The kingdoms and the protozoans. In Parasitic protozoa, Vol. 1., J. Krier, ed., p. 1-34. Academic Press, New York.

g. Whittaker, R. H. Evolution of species diversity in land communities. In Evolu- tionary Biology, Vol. 10, M. K. Hecht, W. C. Steere & B. Wallace, eds., pp. 1-67. Plenum Publishing Corporation, New York.

h. Whittaker, R. H. & S. A. Levin. The role of mosaic phenomena in natural communities. Theor. Popul. Biol. 12: 117-139. Noy-Meir, I. & R. H. Whittaker. Recent developments in continuous multivariate techniques. In Ordination of Plant Com- munities, R. H. Whittaker, ed,, pp. 337-378. Junk, The Hague.

b. Whittaker, R. H., ed. Classification of Plant Communities. Junk, The Hague. 408 pp.

c. Whittaker, R. H., ed. Ordination of Plant Communities. Junk, The Hague. 388 pp.

d. Whittaker, R. H. & H. G. Gauch, Jr. Evaluation of ordination techniques. In Ordination of Plant Communities, R. H. Whittaker, ed., p. 277-336. Junk, The Hague.

e. Whittaker, R. H. & L. Margulis_ Protist classification and the kingdoms of organisms. BioSystems 10: 3-18.

f. Woodwell, G. M., R_ H. Whittaker, W. A. Reiners, G. E. Likens, C. C. Delwiche & D. B. Botkin. The biota and the world carbon budget. Science 199(4325): 141 146.

g. Whittaker, R. H. Review of Terrestrial Vegetation of California, M. G. Barbourand J. Major, eds. Vegetatio 38: 124-125. Naveh, Z. & R. H. Whittaker. Measurements and relationships of plant species diversity in Mediterranean shrublands and woodlands. In Ecological Diversity in Theory and Practice, F. Grassle, G. P. Patti, W. Smith & C. Taillie, eds., p. 219 239. International Co-operative Publishing House, Fairland, Maryland.

b. Olsvig, L. S_, J. F. Cryan & R. H. Whittaker. Vegetational gradients of the pine plains and barrens of Long Island. In Pine Barrens:

1978a.

1979a.

Ecosystem and Landscape, R. T. T. Forman, ed., p. 265-282. Academic Press, New York.

c. Sabo, S. R. & R. H. Whittaker. Bird niches in a subalpine forest: An indirect ordination. Proc. Natl. Acad. Sci. 76:1338 1342.

d. Shmida, A. & R. H. Whittaker. Convergent evolution of deserts in the old and new world. In Werden und Vergehen von Pflanzen- gesellschaften, O. Wilmanns & R. Ttixen, eds., 437 450. Ber. Syrup. Int. Ver. Vegetationskunde, Rinteln, 1978. Cramer, Vaduz.

e. Whittaker, R.H. Appalachian balds and other North American heaths. In Heathlands and Related Shrublands of the World, Vol. 9. A. Descriptive Studies, R. L. Specht, ed., p. 427-440. Elsevier, Amsterdam.

f. Whittaker, R. H. Vegetational relationships of the pine barrens. In Pine Barrens: Ecosystem and Landscape, R. T. T. Forman, ed., p. 315-331. Academic Press, New York.

g. Whittaker, R. H., L. E. Gilbert & J. H. Connell. Analysis of two-phase pattern in a mesquite grassland, Texas_ J. Ecol. 67: 935-952.

h. Whittaker, R. H_ & D. Goodman. Classi- fying species according to their demographic strategy. I. Population fluctuations and environmental heterogeneity. Am. Nat. 113: 185 200.

i. Whittaker, R_ H., G. E. Likens, F. H_ Bormann, J. S. Eaton & T. G. Siccama. The Hubbard Brook ecosystem study: Forest nutrient cycling and element behavior. Ecology 60: 203-220.

j. Whittaker, R. H. & Z. Naveh. Analysis of two-phase patterns. In Contemporary Quan- titative Ecology and Related Ecometrics, G. P. Patiland M. Rosenzweig, eds., p. 157-165. International Co-operative Publishing Hou- se, Fairland, Maryland.

k. Whittaker, R. H., W. A. Niering & M. D. Crisp. Structure, pattern, and diversity of a mallee community in New South Wales. Vegetatio 39: 65-76. Naveh, Z. & R. H. Whittaker. Structural and floristic diversity of shrublands and woodlands in northern, Israel and other Mediterranean areas. Vegetatio 41 : 17 I- 190_

H. G., Jr., R. H. Whittaker & S. B.

1980

1981a. Gauch,

Singer. A comparative study of nonmetric ordinations, J. Ecol. 69: 135-152.

b. Shmida, A. & R. H. Whittaker. Pattern and biological microsite effects in two shrub communities, southern California. Ecology 62:234 251.

c. Woods, K. D. & R. H. Whittaker. Canopy- understory interaction and the internal dynamics of mature hardwood and hemlock- hardwood forests. In D. West, H. H. Shugart & D.B. Botkin (eds.), Forest Succession: Concepts and Application. Springer-Verlag, New York. p. 305-323.

d. Gauch, H. G., Jr. & R. H. Whittaker. Hierarchical classification of community data. J. Ecol. 69: 537-557.

Manuscripts

a. Goodman, D. & R. H. Whittaker. Classi- fying species according to their demographic strategy. II_ A critique of theories concerning r- and k-selection.

b. Whittaker, R. H., J. Morris & D. Goodman. Pattern analysis in savanna at Nylsvley, South Africa.

Table 2. Doctoral dissertations directed by R. H. Whittaker. Degrees were awarded by Cornell Uni- versity, except where indicated.

A. Dissertations completed under R_ H. Whittaker (all at Cornell University)

W. E. Westman, 'Production, Nutrient Circulation and Vegetation-Soil Relations of the Pygmy Forest Region of Northern California.' 1971.

D. C. (Glenn-)Lewin, 'Diversity in Temperate For- ests_' 1973.

T. McHargue, 'A Vegetational Analysis of the Coachella Valley, California.' University of Ca- lifornia, Irvine. 1973.

S. P. Bratton, 'The Structure and Diversity of Harbaceous Understory Communities in Tem- perate Deciduous Forest.' 1975.

R. K. Peet, 'Forest Vegetation of the East Slope of the Northern Colorado Front Range_' 1975.

121

T. R. Wentworth, 'The Vegetation of Limestone and Granite Soils in the Mountains of South- eastern Arizona.' 1976.

O. Sholes, 'Response of Arthropods to the Phenol- ogy of Host-Plant Inflorescences, Concentrating on the Host Genus Solidago.' 1980.

S. Sabo, 'Community Ecology of Subalpine Birds of the White Mountains.' 1980.

L. Olsvig, 'A Comparative Study of Northeastern Pine Barrens Vegetation.' 1980.

B. Dissertations not completed under R. H. Whit- taker

S. R. Kessel,** 'Gradient Modeling: Resource and Fire Management.'

K. D. Woods,* 'Interstand and Intrastand Pattern in Hemlock-Northern Hardwood Forests.' 1981.

M. Wilson,* 'Niche and Habitat in the Low Eleva- tion Vegetation on Serpentine Soils in the Sis- kiyou Mountains, Oregon.' 1982.

References***

Bartholomew, B., 1970. Bare zone between California shrub and grassland communities: the role of animals. Science 170: 1210-1212.

Beals, E. W., 1973. Ordination: mathematical elegance and ecological naivet6. J. Ecol. 61:23 36.

Benz6cri, J. P , 1963 1964. Cours de linguistique math6matique: 3 bme et 4 brae leqons; Rennes. Inst. Statist. Univ. Paris, Multigr.

Bratton, S. P., 1976. Resource division in an understory herb community: responses to temporal and microtopographic gradients. Am. Nat. 110: 679-693.

Bray, J. R. & Curtis, J. T., 1957. An ordination of the upland forest communit ies of southern Wisconsin. Ecol. Monogr. 27:325 349.

Chabot, B. F. & Bunce, J. A., 1979. Drought-stress effects on leaf carbon balance. In: O. T. Solbrig et al. (eds.), Topics in Plant Population Biology, p. 338 355. Columbia Univ. Press, N.Y.

Clements, F. E. &Chaney , R. W., 1936. Environment and life in the Great Plains. Carnegie Inst. of Washington Suppl. Publ. 24 54 pp.

Connell, J. H. & Slatyer, R. O., 1977. Mechanisms of succession in natural communit ies and their role in communi ty stability and organization. Am. Nat. 111:1119 1144.

Curtis, J. T. & Mclntosh, R. P., 1951. An upland forest cont inuum in the prairie-forest border region of Wisconsin. Ecology 32:476 496.

* Not formally completed under R. H. Whittaker. ** Not formally awarded. *** In addition to references listed in Appendix and Table 2.

122

Dagnelie, P., 1960. Contribution b. l'6tude des communaut6s v6g6tales par l'analyse factorielle. Bull. Serv. Carte Phyto- geogr. Ser. B. 5: 7-71, 93-195.

Drury, W. H. & Nisbet, I. C. T,, 1973. Succession. J. Arnold Arb. 54:331-368.

Fisher, R. A., Corbet, A. S. & Williams, C. B., 1943. The relation between the number of species and the number of individuals in a random sample of an animal population. J. Anim. Ecol. 12: 42-58.

Glenn-Lewin, D. C., 1977. Species diversity in North American temperate forests. Vegetatio 33:153 162_

Goodall, D. W_, 1954. Objective methods for the classification of vegetation. III. An essay in the use of factor analysis. Aust. J. Bot. 2: 304~324.

Hill, M. O., 1973. Reciprocal averaging: an eigenvector method of ordination. J. Ecol. 61: 237-249.

Hill, M. O_, Bunce, R. G. H. & Shaw, M. W., 1975. Indicator species analysis, a divisive polythctic method of classification, and its application to a survey of native pinewoods in Scotland. J. Ecol. 63: 597-613.

Hill, M. O. & Gauch, H. G., Jr., 1980_ Detrended correspondence analysis: an improved ordination technique. Vege- ratio 42:47 58.

Horn, H. S., 1975. Markovian properties of forest succession. In: M. L. Cody and J. M. Diamond (eds.), Ecology and Evolution of Communities, p. 196 211. Belknap Press, Cambridge, Mass.

Kessell, S. R., 1979. Gradient Modeling: Resource and Fire Management. Spring-Verlag, N.Y.

Lewin, D_C., 1975. Plant species diversity in ravines of the southern Finger Lakes region, New York. Can. J. Bot. 53: 1465-1472.

MacArthur, R. H & Wilson, E. O., 1967. The Theory of Island Biogeography. Monogr. in Population Biology I. Princeton University Press, Princeton, N. J.

Merriam, C. Hart., 1890_ Results of a biological survey of the San Francisco Mountain region and desert of the Little Colorado in Arizona. North American Fauna 3:1 128.

Noy-Meir, I. & Austin, M. P., 1971. Principal component ordination and simulated vegetational data. Ecology 51: 551-552.

Orl6ci, L., 1966. Geometric models in ecology. I. The theory and application of some ordination methods. J. Ecol. 54:193-215.

Peet, R. K., 1978a. Latitudinal variation in southern Rocky Mountain forests. J. Biogeogr. 5: 275-289.

Peet, R. K., 1978b. Forest vegetation of the Colorado Front Range: patterns of species diversity. Vegetatio 37: 65-78.

Peet, R. K., 1981. Forest vegetation of the Colorado Front Range: composition and dynamics. Vegetatio 45:3 75.

Pickett, S. T. A., 1976. Succession: an evolutionary interpretation. Am. Nat. 110: 107-119.

Phillips, J., 1935. Succession, development, the climax, and the complex organism: an analysis of concepts. Part III: The complex organism: conclusions. J. Ecol. 23: 488-508.

Runkle, J., 1979. Gap phase dyamics in climax mesic forests. Ph. D. dissertation. Cornell Univ., N.Y.

Sabo, S. R_, 1980. Niche and habitat relations in subalpine bird communities of the White Mountains of New Hampshire. Ecol. Monogr. 50: 241-259.

Shreve, F., 1915. The vegetation of a desert mountain range as conditioned by climatic factors. Carnegie Inst. Washington Publ. 217: 1-112.

Smuts, J. C., 1926. Holism and Evolution. Macmillan, N.Y. Swan, J. M. A., 1970. An examination of some ordination

problems by use of simulated vegetational data. Ecology 51: 89-102_

Tansley, A. G., 1935. The use and abuse of vegetational concepts and terms. Ecology 16: 284-307.

Tobey, R., 1981. Saving the Prairies: The Founding School of American Plant Ecology, 1895-1955. University of Cali- fornia Press: Berkeley and Los Angeles.

Vestal, A. G., 1914. Prairie vegetation of a mountain front area in Colorado. Bot. Gaz. 58: 377-400.

Vestal, A. G., 1917. Foothills vegetation in the Colorado Front Range. Bot. Gaz. 64:353 385.

Wentworth, T. R., 1981. Vegetation on limestone and granite in the Mule Mountains, Arizona. Ecology 62 469-482.

Westman, W. E., 1975. Edaphic climax pattern of the pygmy forest region of California. Ecol. Monogr. 45:109 135.

Westman, W_ E., 1978. Patterns of nutrient flow in the pygmy forest region of northern California_ Vegetatio 36: 1-16.

Williams, W. T, & Lambert, J. M., 1959. Multivariate methods in plant ecology. I. Association-analysis in plant communities. J. Ecol. 47:83 101.

Woods, K., 1979. Reciprocal replacement and the maintenance of codominance in a beech maple forest. Oikos 33: 31-39.

Accepted 30.10,1981.

robert h. whittaker (1920–1980): the man and his...

Documents